Nate Habermeyer, APR: 0:00

Hello, listeners. This is the Risky Planner podcast. Thanks for tuning in. Hey, Albert. Hey, Nate. How are you doing? Oh, you know, I've I've been running all over creation, doing all kinds of different things. It's been an extremely busy month for me. It has been, and we're going to talk about AACE and this paper that you presented, as well as some pretty cool, yeah, HR conversations. We are 100% Like I'm really excited to talk about this, and I have been excited to talk about it for a month since it is a month ago that we actually went to the AACE conference. A month ago, Nate. I know. I mean, now you weren't there this time, which was unfortunate. The show and the conversations that you had-it sounds like it was really productive, and so yeah, I'm a little disappointed that I wasn't there. But I know you guys had a good time, and it sounds it sounded very productive. And it like we'll talk about some of the conversations you had.

Albert Brier: 1:15

Yeah, but before we get into that, yeah, you mentioned the paper we wrote, and this led because I went with with your friend and mine, Roger, star of the last episode, and we spent a lot of time in technical sessions. We spent almost all of our time in the technical sessions. Now that's unusual for me. I don't, you know, I shouldn't really admit this, but I don't have a whole lot of patience for the the technical sessions a lot of the time, but this year's lineup was excellent, and there was a room that was just devoted to risk, and there were a lot of topics that felt like risk that were actually grouped under other categories, you know, and so we wound up attending project management sessions, communication sessions, estimating sessions, scheduling sessions, risk sessions-we were in technical sessions basically all the time.

Nate Habermeyer, APR: 2:07

Did you have any favorites that you went to that you thought were particularly insightful, or that you would like to share here?

Albert Brier: 2:13

Yeah, I mean, there were there were a few favorites, but I'll point to some general trends instead. Basically, some people doing really interesting things with data visualization in risk space. One of the things that happens is you generate a lot of data, and it's multi-dimensional data. So whenever you have multi-dimensional data, that immediately makes it difficult to parse, right? Like you know, we we operate we humans operate pretty well in two dimensions. Like that, that's pretty intuitive for us. You look at a screen, you understand what's going on. Kids draw pictures, two dimensions, right? Yeah. In three dimensions, we can wrap our heads around it because we live in a three-dimensional world. You know, we can interact with things in three dimensions. We understand the difference between the various axes and all that stuff. But as soon as you go above three, it gets really hard to visualize, right? So that's when we start to bring things in, like colors, right? Where like the the a fourth dimension of data might be a color, or or it's like a a different like a contour line that runs through three dimensional data, right? So were

Nate Habermeyer, APR: 3:12

you seeing visualizations that were kind of like taking that? Okay, interesting. So this kind of reminds me of that digital twin conversation that we had, right? Like talking about like 4d, 5d, like you're just layering, layering on. It

Albert Brier: 3:26

is a it is different stuff, as I like to say. It's kissing cousins to that because it it does have kind of like a a data visualization layer onto something that might feel like it should be intuitively obvious what it means because like people like to smash down risk into how long it's going to take and how much does it cost, right? Another thing that was kind of a running thread through the risk discussions is people are trying to reframe risk as decision support, which like

Unknown: 3:50

okay, same

Albert Brier: 3:51

rebrand, yeah. So if you think about risk as as like that that smash down concept I just talk about, where it's how much is it going to cost and how long is it going to take, when, right? It's a moment in time analysis. Most of these are right. There's very there's very few products, maybe none products, that will actually do an ongoing accounting of your risk exposure over time. And if they do, they're not doing it with like a fully integrated Monte Carlo simulation, right? They're just doing it with you know say values from your risk register or something that maybe hand entered into like your scheduling tool, like p6 or something, right? So they're not the same level of rigor. It's not the same thing. So all of those big heavy analyzes are moments in time. So why have one? And this is not a trivial question because it's one that executives ask all the time. Like if you want to do a really good job doing like a full-blown Monte Carlo integrated risk assessment for a mega project, that's probably a $250,000 touch, maybe more. Okay, right? Yeah. So you know that could be a team of 1015, people working for weeks to get something like that done, and that it. Doesn't necessarily make sense to the average person that it would be that that much effort to do, but you have to gather a lot of information from a lot of people. So the the trick to reframing risk as decision support is it takes the focus off of the numbers and instead puts it onto the data that was captured and how can you flow that data through the project such that it reaches the ears of executives who need to make decisions about it. Because if you can tell somebody, well, your team is saying this and that and this other thing, and they didn't have their arms around the schedule, but now they have a better idea. We looked at some of the analogous projects and how they ran their things and how long we actually did a lot of this with with a recent client framing up the results in the context of here's where your project sits in the broader project ecosystem, like across the world or in your category of of company, and here's what the number says. Right, right. So you take all those things, you put the number very, very, very last, and you say instead, hey, you know, executive, and I already know this because you're so smart and handsome, but here's the information that you need to make the decision about whether this project goes forward or not, because it isn't going to cost X; it's going to cost 4x. But here's all the very good reasons, and that actually puts you in really good company. Yeah. Right.

Nate Habermeyer, APR: 6:11

Yeah.

Albert Brier: 6:12

So it's the difference between like here's a histogram and here's some context.

Nate Habermeyer, APR: 6:15

Like I hear that and I think, oh, you know, AI and you know other tools exist now that expedite sort of the crunching of data or visualization or or just enabling professionals in the capital project space to synthesize risk and then report on that faster. But we'll get into some of the interviews, and and it doesn't. But it doesn't sound like there's a lot of AI in use, but it's just what I heard was like, it seems like there is more help, but I don't know. But it's still really so interesting. Yeah, it

Albert Brier: 6:52

was like when we get into the actual interview content, that was one of the trends that emerged for me that I found the most interesting was was how people are actually using AI for things like this, but before we get into that, I will say you know Roger was sitting next to me in the technical sessions with his laptop open, using AI to regenerate some of the visualizations and some of the concepts that were being presented in the papers in real time. Oh, so he'd have the paper open, yeah, and like a Claude window open, and be working in Claude code to just like jam out ad hoc HTML visualizations for things. Yeah, that's cool. It was really cool. I mean, okay. So there's one that might actually find its way into a software product near you not too long from

Unknown: 7:34

now.

Albert Brier: 7:34

But anyway, it was it was very neat to to kind of have that extra lens on it. And of course, you know, Roger is the gearhead amongst us, and he just as soon as he saw people like, hey, here's a cool computerized visualization could do, you know, his brain immediately went to, well, I'm going to do that right now.

Nate Habermeyer, APR: 7:51

So was he taking their papers and improving on what their visualization might be, or was he kind of putting his own spin on their, he was largely just

Albert Brier: 8:03

implementing them. You know, just taking what was shown in the paper and throwing some dummy data at it and saying, "Well, how could I actually implement this in a live way? Right, because like a lot of the things you're seeing in the paper, obviously they're static; they're in a paper. But the things that you're seeing in the paper are, you know, the results of weeks of analysis, right? So if you take a live dataset and want to reproduce that instead of in weeks in minutes, you know how would you do that, right? That was the kind of thing that that Roger was doing and playing with. He built one that was one of those 4d visualizers I was talking about earlier, like a color coded 3d visualization that runs in real time, butter smooth, takes custom data as inputs. It was it was wild to see that come together in like less than 45 minutes. But it's interesting,

Nate Habermeyer, APR: 8:51

like you're talking about, there is analysis that people do, and then there's the execution of that analysis to make it consumable. Like that, there's something to be said about that making something consumable, yeah, yeah,

Albert Brier: 9:03

and even just like somebody pointing the way is often all that's required to democratize it, right? Right. Like the first person to climb Mount Everest was somebody who lived on the slopes of Mount Everest their entire life and was acclimated to the climate and thought, "Gee whiz, it'd be interesting to climb it. Like that was probably the first through 100. I'm sure that's what they said

Nate Habermeyer, APR: 9:21

too. Is like I'm gonna climb that

Albert Brier: 9:23

nothing. I'm just gonna go climb that. Why not? But like you know, the the first person ever get famous for it was like a Westerner who did the same thing, and that took like you know years and years of planning, and you know people died, and it made the news all over the world, and this person became a celebrity, and the second person kind of the same thing. The 10,000th person to do that is a tourist from Ohio, right?

Nate Habermeyer, APR: 9:45

I was born in Ohio.

Albert Brier: 9:47

Have you been to Mount Everest?

Nate Habermeyer, APR: 9:49

No, no, that's yeah. But you could no. You know,

Albert Brier: 9:52

my my point though is that there's a world full of practitioners who live and breathe this stuff who are probably doing things like this. Yeah, and then somebody goes and writes. Paper about it, and that's you know your your Western explorer summiting Everest for the first time, and then the the gap between that and the Ohio tourists is just getting a lot narrower. Shout out to Ohio. Shout out to Ohio. Yeah, yeah, I love your tourists. You guys do Everest real good.

Nate Habermeyer, APR: 10:17

Tell me about your paper. Like, okay, those were cool. Let's switch gears. Like, how was your presentation?

Albert Brier: 10:22

Our presentation, I thought, went really well. the The paper received top marks from from the review committee, which was nice to see. There was a lot of excitement in general about the content of the paper. What we were trying to do was find a way to standardize program risk, and we talked about that in detail in the last episode. By the way, one thing that we owe our listeners is a quick primer on Monte Carlo simulation. We got some feedback from the last episode that we were we were talking as if everybody already knew what it was and how it functioned, and maybe that's not true. So we'll we'll do a quick one on that to talk about what what that is and how it works. But anyway,

Nate Habermeyer, APR: 11:01

yeah, kind of like a short, you know, like a playbook.

Albert Brier: 11:03

Point is, our paper takes Monte Carlo simulation at the program level and gives you a framework for doing that without overstuffing a risk model with every single schedule for every single project in your in your program. It allows you to identify risk at the sub project level and then gives you like a credible, mathematically valid framework for rolling it up to the program level. We are the first group of people to document a way of doing that that is standardized and repeatable. That does not mean that we're the first group of people ever to do it. Okay, it just means that we have a way that's now published and has the AACE logo on it.

Nate Habermeyer, APR: 11:40

Yeah. Right. Yeah.

Albert Brier: 11:41

Which means that the next step is going to be to write an RP about it. We got a lot of good feedback from the DRM subcommittee about this paper being pretty much ready for RP writing. So we're going to take the next step pretty soon here of of building the draft RP and getting it in front of people's for the

Nate Habermeyer, APR: 11:58

recommended practice. Is that yeah

Albert Brier: 12:00

RP? Yeah, the recommended practice. And then,

Nate Habermeyer, APR: 12:02

what? Tell me about questions or feedback. Like, people obviously said,"Oh, it's good. Like, do you remember any of the detailed feedback? There were

Albert Brier: 12:10

two things that I got questions on pretty regularly that that I thought were really interesting, and one that was like a total, just utterly practical, but that I just had not thought of before. Okay,

Nate Habermeyer, APR: 12:20

yeah, that's

Albert Brier: 12:21

cool. So the two things that came up the most were G Wiz. This sure is complicated. What organization on the planet could possibly execute this? To which I have to say, I don't know. Like it's a really good question. Jokanish and

Nate Habermeyer, APR: 12:33

Company could execute it.

Albert Brier: 12:34

Well, we certainly could. That's right. the The thing that needs knowing though is that Jokanish and Company we're we're really smart, talented people, and we do a lot of really cool things with risk. But we don't own things, right? We don't own nuclear generation stations. We don't own highway networks. You know, we we don't own submarines. We own our processes, and we own our intellectual property. But we don't own the actual assets that need this risk work to be done in support of their construction or redevelopment. So we need to. There is a high bar for for for folks like us to clear in order to get this stuff in front of the people who really need it. Right. Hence, again, like reframing this all as decision support.

Unknown: 13:19

Yeah.

Albert Brier: 13:19

You know. So that was one thing that came out. The other one was people had lots of questions about schedule, like how schedule fits into this picture. How schedule? Okay. So we that is a very good question in in part because you know cost is additive, schedule is not. Even in risk space, cost no longer becomes one to one additive, but it does still build up, right? In in a way that feels logical when you see it graphed out, schedule just doesn't. Right? Schedule is is really weird when you're trying to add schedule values together. You know, even in deterministic space, one plus one does not equal two. That's true in stochastic space for cost values, but it's true in deterministic space for schedule. Right. So that that makes schedule much much harder to integrate into a program level recommended practice about how to manage contingency, for example. But it's something that I want to write a paper about for the upcoming conference because I want to test drive some ideas. Not quite sure how I'm going to frame that one up yet, but it's it's it's brewing. It's something brain. Okay, interesting. Yeah.

Nate Habermeyer, APR: 14:22

Anything else?

Albert Brier: 14:23

Yeah. There's that left field one I wanted to tell you about. I spoke with a gal who actually lives not too far from me who asked one of those practical questions that needs to come up during the RP review, which is, okay, so you're saying you need a thick layer of management at the program level. You want to maintain some budget at the program level for doing you know risk paying for risk, right? Like the management reserve layer at that P 80 value in our paper. That's that's the way we framed it up. And she said, "Well, okay. So what if the program is being executed by a bunch of different organizations who don't share a budget? That's a really good question.

Unknown: 14:57

Yep.

Albert Brier: 14:57

So we sat there and kind of spitballed. That like between the two of us, we came up with an answer that would work. Oh yeah! But it's the kind of thing that maybe needs a little footnote in the RP. Like, look, if you can't hold a budget, right, then you can virtualize a budget. You know, you can lock out decision authority for that budget from the project managers. Like, give it explicitly through process, you know. Maybe you can't engineer that control into your financial system, but you can put a process control in place to say like your spending authority goes up to you know 92% of of your PO limit or whatever.

Nate Habermeyer, APR: 15:33

Okay, would this be like a consortium of of different companies coming together to build? She works in utilities,

Albert Brier: 15:39

so it's it's actually it kind of is because there's the municipalities across you know Canada is a very large country. That's

Nate Habermeyer, APR: 15:48

one way to put it

Albert Brier: 15:49

with relatively few people in it. So a lot of the the infrastructure projects that that get built here have a much higher per capita cost than they would in a in a larger country. So how things like power lines and highways get built, the costs get split up between a bunch of different groups of people, and that means that there is no one budget for the whole program. Right, right. So that was an interesting conversation for sure.

Nate Habermeyer, APR: 16:17

All right, well, let's talk about some of the conversations that you had on the floor, because yeah, let's do it. They are pretty interesting, as we alluded to when we talked about the AI.

Albert Brier: 16:29

Yeah,

Nate Habermeyer, APR: 16:29

last year we did the same thing, right? Like you went to AACE, you had a bunch of interesting conversations. You interviewed some of the same people this year that you did last year, or maybe one or two, a few repeat performances. Yeah, tell me, like overall, what was that like? Approaching people on the floor and asking for interviews, and you know, what what did you learn?

Albert Brier: 16:52

People were really into it. It was it was very easy to have the conversation this time around because I didn't feel like I was doing something unusual. I was doing something unusual, but not to me. I'd already done it once, right? And some of the people who I interviewed, they were more comfortable with it as well. So it was it was a very easy conversation to have. The only thing that made it hard, and the only reason I didn't collect more interviews, is because as I mentioned previously, Roger and I just spent so much time in the technical sessions. But we did get a lot of interviews from immediately after those sessions with the people who presented papers, so it's a different mix of people this time too. Yeah, you know, fewer software vendors this time. We spoke to basically every vendor on the floor last year, and like I actually got an email from the from one of the Oracle reps saying like, "Hey, you didn't come talk to us. Yeah, sorry. Nice. Yeah, just just slipped my mind. I'm so sorry. So, but we we tried to stick with practitioners, and we we looked for more people who work for owner operator type of organizations, or people who own that infrastructure I was talking about before. Like we we got a lot more input from those folks than we did last year, which was fun.

Nate Habermeyer, APR: 17:58

And the questions that you asked,

Unknown: 18:00

yeah,

Nate Habermeyer, APR: 18:01

there were some really interesting trends to that emerged. Right? Who did you interview on the floor?

Albert Brier: 18:08

Yeah, it was a bunch of different kinds of people. But why don't we just let them introduce themselves?

Nate Habermeyer, APR: 18:12

Okay, roll the tape.

Unknown: 18:14

I am Garish Batia. My job title is I am the founder and CEO of ConstructMine AI. Justin Jacobson. I'm the vice president of innovation development for MVP. Matthew Schoenart. I'm a senior risk principal. I'm with Hatch Consulting. Hi, I'm Michael Lapage. I'm the CEO, founder of Plan Academy. We're a training company with a big focus on scheduling delays, and we do some consulting too. Philip Talbot. I'm an enterprise account executive for Nodes and links. My name is Prakash Hundeker. I work for Volker Inc. and my position is project controls manager. Bear Richards, project director with the U.S. infrastructure team at Turner and Townsend. My name is Belinda Hassey. I am the cost engineering manager for Transpower New Zealand. Diana Nada. I'm a director at Turner and Townsend in the advisory team. My name is Holly Parkus. I'm the vice president of advisory services for SMA Consulting, and we're based in Edmonton, Alberta. Rachel Fleming. I am the service line manager for risk and facilitated services at MBP. I'm Rhys Phillips. I'm a principal deployment engineer at Mplan. Rohit Sena. I'm the CTO and one of the co-founders at SmartPM Technologies. I am Sarah Horsey. I am the project. I am a project controls director for Shakun and Benue USA Construction, and we are currently serving as the project controls manager on the Major Bridges P3 initiative. Tracy Lung. I am currently the vice president of Environment, Health, and Safety at Ontario Power Generation. Very

Nate Habermeyer, APR: 19:47

interesting group of people.

Albert Brier: 19:48

Yeah, good mix, right?

Nate Habermeyer, APR: 19:50

Yeah, very good mix. What was your kind of takeaway? What did you sort of learn, or how did you feel after talking to all those folks?

Albert Brier: 19:57

Well, you know what? I think this is going to be an easier. To answer after we talk about the next question, which was what secret sauce do you bring to projects? That was the way we literally phrased it. So let's let's hear what people had to say about themselves.

Nate Habermeyer, APR: 20:10

All right,

Unknown: 20:11

we are working on crunching the schedule generation time from 25 days to 25 minutes using artificial intelligence, a structured way to identify and collaborate on risks in order to benefit the project. I have done a quarter trillion dollars of quantitative risk analysis, so I have seen a lot of projects and a lot of project teams. I don't know if I bring this to projects, but one of the things I like to do is to just share information. So we have a popular YouTube channel. We make lots of videos. We want people to understand some of the weird nuances of scheduling or working, say, with P6 or Microsoft Project, and I think it should be widely available for all people. For us, it's a mixture of project knowledge, but it's also bringing AI to the scheduling space from an analytics perspective. My secret sauce is pure grit, determination, experience in the field for 20 plus years. Probably my dad jokes. Can you tell us a dad joke right now? Where do broken down Volkswagens go? I don't know. Where do broken down Volkswagens go? The old Volkswagen. I would actually like to say it's about being a bit more of a generalist, and I think that's really important because you can drop in and drop out of so much more than the deep dive into being a specialist. Part of my research was around behaviors and psychology of major problems, so I see that psychology piece is something that I think we're struggling with because projects are people, and we're not improving on how we deliver. So I think we need to see why we do what we do all the time. It's risk, I would say. I think we view everything through the lens of risk, and that drives all of our other work. Focusing on how information is messaged and where and how communication flows between stakeholders within a project. Secret sauce is the ability to leverage the world's largest dataset of historical project outcomes and machine learning models trained on them to predict on current and future projects to understand outcomes. Ah, the secret sauce we bring to projects is we bring everyone together on the same data, so that everyone is on a level playing field, and that's really what leads to better execution. I bring the human side of projects. Projects isn't always about getting the pump installed and getting the concrete poured and or costs and schedule. There are people receiving the benefits at the end of the project, and there are people executing projects. I think that side, people side, always fascinates me, and I enjoy addressing that side of the project.

Nate Habermeyer, APR: 22:48

That was pretty cool.

Albert Brier: 22:49

Yeah, you know what stuck out to me was a lot of people felt like their secret sauce was interpersonal connection and communications. That's right. That seemed to be the emergent trend from that. And I didn't ask this question last year, but I sure wish I had because I would love to know what the difference would feel. You know what I mean? Like, are people saying that more this year than they were last year? Because you know, last year we predicted that this year would be the year when AI really takes hold and people start, you know, using it heavy duty. And I think that has happened. You know, I I read a really interesting article, like an opinion piece, about how taking a lot of the really heavy-duty lifting away from your average office worker means that the the the skills that are the most valuable to an organization are now those interpersonal skills, right? Like the ability to have a difficult conversation, the really the ability to package a difficult concept up; those skills are the ones that survive the the great filter that we're currently going through right now. So it's it's no surprise to me that people are talking about

Nate Habermeyer, APR: 23:51

it. Yeah, I think no matter what kinds of technology are introduced or new processes or best practices, communication and that interpersonal relationship is going to always be key, right? To doing a really good job.

Albert Brier: 24:07

It feels like this is just hard to predict right now because of how fast things are moving. But it feels like one of those things that would be very difficult, if not impossible, to ever replace with automation meaningfully. That's right. You know. That's right. So somebody's going to have to go into a room and talk to a group of strangers about what they do for a living, and that is just the the that's going to be true forever, right?

Nate Habermeyer, APR: 24:30

Yep. Yep. Well, it's a good segue because your next question was about AI, and this was particularly fascinating for me to listen to, and a lot surprised me. Yeah. So this

Albert Brier: 24:42

one was my favorite. So let's let's take a listen.

Nate Habermeyer, APR: 24:44

Roll the tape.

Unknown: 24:45

Our AI is designed to convert project documents to a schedule that is with logics, durations picked up from first principles, resources assigned, and if you have cost embedded into your resources, you could have. Cost-loaded schedules as well, leveraging AI to help us develop discrete software of our own to solve certain problems. Nothing really. Although I've done some great talks here, and I'm thinking I should be incorporating some. I'm trying to figure it out. Like a lot of people, we're allowing people to access their schedule in a way that they haven't ever before, where you can probe your schedule and ask questions that are natural language questions, trying to figure out anything from what's changed to what risks do I have coming up. Well, we just started recently with AI in our company, so we got a subscription for Claude, and I am tasked to develop some projects. So we'll see how it goes. Yeah, I'd say it's it's pretty light, but it is it's becoming more and more a regular tool that I'm using. Doing a little bit in my team, we've developed a couple of agents which we're using, but only to create output templates from a consistency point of view. Not much personally at the moment, just like ChatGPT, etc. We want AI to be a tool rather than the driver. I'm still exploring and trying to understand where and how AI can be helpful, value-added, where it can help with efficiencies, and interested to see where the future takes us. We've been building, training, deploying models in the real world for nearly a decade now. We work with some of the largest infrastructure developers, owners, operators, and contractors around the world. We know that our data is calculated correctly because we've designed and battle tested those systems. So AI is really just we're leveraging it to do the things that it's very good at, right? Like converting our analytics over to text, but then also custom classification models. We're also leveraging it to link between different data sets. We're using it a little bit on the project level for like consolidating safety statistics and metrics, but we haven't really started implementing it on the project control side. More on the administrative side of things. Like we rely on AI, for example, in my role to check on compliance to law and and regulations,

Nate Habermeyer, APR: 27:07

a lot of surprises there. Some trends, Albert.

Albert Brier: 27:09

Great. Yeah. So the world divided in two for that big time for that question, and I'm gonna I'm gonna put that those two groups are people and software companies, right? Right. So you know, people ran a very small spectrum between. I'm not really using it at all. To I'm kind of test driving it in in. There's a lot of hesitancy, right? And I was surprised

Nate Habermeyer, APR: 27:37

by that at this point, like at this juncture, at this time, we're midway through 2026. I'm just surprised, but it actually doesn't. I mean, if you think about sort of the corporate adoption of AI and the return on investment, and there's a lot of conversation in the news, are that it's actually not surprising.

Albert Brier: 27:54

Yeah, you have to think about it from two, from like a push pull perspective. Like people are being pushed towards adopting AI, and even the people that I spoke to at the conference, who you know, quote unquote, aren't using it, they were mostly talking about their technical work. Everybody uses AI all the time for for a variety of things, whether they want to or not. Like you Google something, you're using AI, right? Because you get the, and that's been true for a long time, even since before they started branding it as like AI summaries. There was always an artificial intelligent machine learning component to to most things that people do on the internet, right? That's

Nate Habermeyer, APR: 28:27

right. So put

Albert Brier: 28:28

that aside and think just about the technical work, and it does start to make a lot more sense. Like you mentioned, I think very smartly that organizations are not seeing the return on investment from AI adoption that they thought they were, and some of the companies that fired a bunch of people thinking they could automate all those tasks have have started rehiring because they turns out they can't right yeah there's a now they didn't rehire everybody so there's there's a floor to how much of that capture was was overestimated but it was some right and when you hear these highly technical people talking about this brings me to think too. These these are people largely who have been doing this work for decades, right?

Nate Habermeyer, APR: 29:05

Yep. Yeah.

Albert Brier: 29:06

Who who really really know what they're talking about? Who are experts in their tools, experts in their fields. They've been there, done that. Why would they trust an AI? Why?

Nate Habermeyer, APR: 29:17

Yeah. There's a lot of mistrust in those answers. Like a number of people talked about trust.

Albert Brier: 29:22

Yeah, yeah, and like we there's been no push within Docanish and Company to like thou shalt use AI. Like we're we're not being measured on our use of AI like some people are, but a lot of our people have started adopting it and using it voluntarily, and kind of cautiously, optimistically test driving it for doing certain things. Roger and I are amongst the heaviest users in that respect. We're really trying to push the envelope of what a technical expert can use AI for, and both of us have had instances where you you try to use it for something and it's a little bit too heavy a lift for the thing, but you don't realize it until you've already spent a bunch of time trying to correct it and. And so it becomes like, was it a wash? Like, could I have just done this myself? Could have

Nate Habermeyer, APR: 30:04

just done it, and it would have taken you less time.

Albert Brier: 30:06

Yeah, and now that's not true for everything, but but the things that it is true for are not clear, right? It's

Nate Habermeyer, APR: 30:13

true.

Albert Brier: 30:13

So, and some of the really heavy duty technical stuff, things like planning and scheduling, right? I know from a lot of personal recent experience, that AI just is not there when it comes to assembling a logic network, estimating activity durations. You know, you can put together a really brilliant activity list for you, but it'll overstuff a WBS like really mightily overstuff a WBS. There are just some things that it's just straight up not good at, and you're not going to figure out what those are unless you've test driven it like a lot, like I have, but a lot of people just don't have time to to spend on that kind of. It does take

Nate Habermeyer, APR: 30:48

time, yeah, to set it up. It is not a turnkey solution. It takes a lot of time.

Albert Brier: 30:53

No, unless you're trying to polish an email or something. Was that's super low value to the kinds of you know technocratic folks I was speaking to at the group or at the conference, and then there's the companies, right? The software companies, yeah. And they are like 15 out of 10 bought into AI. We got an AI that'll do this and that and everything, and we have a 7 million data point data set, and we got all this stuff, and we're doing all these things, and you know companies are excited. We have 10,000 clients and all that stuff, and it's like all the people who work for those companies are saying they are either doing nothing or very little with AI, but you're selling the 15 out of 10 ultra deluxe spit polish platinum edition AI bundle. Like that's that's what the that's what it felt like. You know, walk on the floor.

Nate Habermeyer, APR: 31:37

Yeah, the practitioner is very cautious and distrustful of whatever that solution is, and then the software vendors have this. I've got the turnkey solution. There's there's a gap. Something's not connect connecting, but very interesting.

Albert Brier: 31:51

I found that divide super fascinating because I think it it told me everything that I didn't yet understand about like where AI is in the world today, that's what told me that. You know what I mean? Like I, I got, I, I got it after talking to all those people because I'm like, oh, all right, this, this is what this is what's going on on the internet when like random folks are saying they don't trust it or they don't use it or they wish nobody was using it or they're like really mad at it or they're really confused why there's all this investment in it and that sort of thing. This is what it is. This is it at the ground level. It's what it looks like.

Nate Habermeyer, APR: 32:26

Yeah, I was really.

Albert Brier: 32:27

I'm not throwing any shade at the software vendors, by the way.

Nate Habermeyer, APR: 32:29

Thank you. The stuff

Albert Brier: 32:30

that they are doing is very, very cool, and it's just interesting that it's be it's it's a supply driven exercise all across the spectrum, even within the the companies that make an AI product, like Meta, for example, they're they're pivoting their AI infrastructure spend from software development to data center construction. That's

Unknown: 32:51

right,

Albert Brier: 32:51

right. Like there was a recent large, like a mega project, in fact, identified within Alberta, right, from Meta to build a data center in the Edmonton area, $13 billion dollar project, supposedly. So that is Meta, the same Meta that makes Instagram and Facebook and stuff, and has their own AI product. They're building data centers now, and that may sound like well, of course they are, but like it's it's not an of course there. Like they're not the kind of company that you would typically have seen building their own massive data centers. Just a few years ago, they're not Amazon. They're not Microsoft, right? They're not Oracle, and and then you've got these software vendors who are saying we have all this processing power at our at our fingertips. We have all this historical data. Let's use it, and they're and they're using it in ways that are really really cool. But it needs to start trickling down to actual end users. But isn't that isn't that

Nate Habermeyer, APR: 33:42

how how innovation kind of works? I think these tech companies and software companies are very forward thinking, introducing new solutions, and there's a there's going to be catch up for the practitioners and industry to start adopting some of these cool New solutions, and then others will be sort of weeded out, right? Like that. That it's the Gartner hype cycle, basically.

Albert Brier: 34:09

Yeah, no, it it it is. But I want to compare it to something like a technological innovation that we're we're all very familiar with, which is the the the car, right? The personal automobile. The personal

Nate Habermeyer, APR: 34:20

automobile. The

Albert Brier: 34:22

personal automobile. Yes. Okay. That's what I like to call them. What is your personal automobile, Nate?

Nate Habermeyer, APR: 34:30

I have one personal automobile. That's your bike. I have technically

Albert Brier: 34:32

an automobile.

Nate Habermeyer, APR: 34:33

It's my whip.

Albert Brier: 34:33

It's your whip.

Nate Habermeyer, APR: 34:34

So a personal automobile.

Albert Brier: 34:36

So the first one of those was a bespoke, one at a time, handcrafted in a factory in Germany for rich people, kind of thing, right? That's the first cars. Even the first cars in America were high-dollar items, and and that that happened like 20 years before the first mass-produced car. Like, and and you and I, you know, both American kids, we we grew up a certain way. What was. First mass-produced car. Just say it out loud.

Nate Habermeyer, APR: 35:02

That's a good. Oh,

Unknown: 35:03

Model T. Yeah, the Model T. Of course. Okay, I don't.

Albert Brier: 35:06

You were looking for the hard answer, but it was the easy. I was like,

Nate Habermeyer, APR: 35:09

yeah, I was looking for the hard answer. I was like, Ford Taurus

Albert Brier: 35:13

was it the Mercury Sable?

Nate Habermeyer, APR: 35:14

No. Yeah, exactly. I don't know why. I think

Albert Brier: 35:16

it was the Buick Grand National. But anyway, yeah. What a

Nate Habermeyer, APR: 35:20

car!

Albert Brier: 35:21

Yeah, right. The GNX. Oh yeah, buddy. Yeah. But anyway, the the first mass produced car was the Model T, and it came about 20 years after the invention of the automobile. That's right.

Unknown: 35:30

Right.

Albert Brier: 35:31

So what if instead the first automobile had been the Model T, and right, and there were 1000s of them being produced every year, and they were cheap enough for people to afford, but nobody had that context for what they were for or how to use them. Right,

Nate Habermeyer, APR: 35:46

right.

Albert Brier: 35:46

That's where we're at with AI right now. Yeah,

Nate Habermeyer, APR: 35:49

mind blown. I

Albert Brier: 35:50

know, right? We jumped straight to Model T's. Yeah, skipped. No, but that's

Nate Habermeyer, APR: 35:54

a great analogy because it's it is you know the the softwares are. I think I think there's there's so much potential, but people

Albert Brier: 36:01

there is, and the software vendors are doing a really good job of of putting that rubber to the road, and I think they deserve credit for that. But the the world needs to catch up to them. People need to understand what it's for. Like it's easy to forget that the first operational ChatGPT model is is like less than 10 years old, like a lot less than 10 years old, right? So we're we're not even at the midpoint in that German bespoke automobile for rich people to Model T spectrum. We're not even halfway through there yet, right? Like airplanes, it was a similar thing. Like you couldn't buy a commercial ticket on an airplane, you know, if the the the instant the Wright brother brothers flew for the first time, like that wasn't how it worked, right? The first the first airplane flight was like, oh my god, look at these dudes! They built a flying machine. That's crazy. Now what are we going to use it for? And everybody's like, I don't know. It's like we already got we've already got cool ones. Yeah.

Nate Habermeyer, APR: 36:53

So, but you're you're talking so the big big projects, right? And this is it's a good lead into like your next question, which was also really interesting to hear people's responses, and that was like, tell me about your favorite project, like any project, so not just the one that you worked on, but some people, it was interesting to see or to hear how they answered that question because a lot of people it was about their work. It was about their

Albert Brier: 37:19

work, so let's talk about that in a second. Let's give it a listen.

Nate Habermeyer, APR: 37:21

Roll the tape.

Unknown: 37:22

Oh wow. Okay. So I think the oil and gas industry and the data center industry. These two industries are very close to my heart. I'm working with the Smithsonian Institution on the original headquarters, what they call the castle, and we're doing a base isolation project. So we're digging out the foundation underneath a historic building and putting base isolators under it to protect from earthquakes, which is a really interesting project for a very significant structure in U.S. history. I'm really excited about working with Vancouver Port Authority and Robert Banks Terminal Two. I've been working with them for a couple years now, working with BHP on Janssen One and Janssen two, the project I keep thinking of is the Burg, the Burghalifa. I would have loved to have even a small par, any association with that giant monstrosity of a tower, anything. I would have been amazing to just say, hey, I did you know I did something on that, or I was part of it. Data centers obviously are happening all over the U.S. and that's a really exciting place to be. The data center market is is really taking off, and those owners and contractors are operating at a different level than what we've seen in the past. Highway project in Austin, they first design bid build major project, Central Texas Turnpike. It was close to a billion dollar, and I was working on a$200 million interchange at SH 45 US 183. I would say the creation as recorded in the Old Testament of the Bible. Well, in my current role, it's about our HVDC project, which is our big upgrade for our cables between the North and the South Island, so we can make sure that we are up to speed for things that are coming to end of life. The longest that I've worked on was something was the waterfront Toronto development, and it was setting up the PMO for this big development and the flood protection in the Toronto region. In the next 20 years, basically, it will change the skyline of Toronto. I'm from Massachusetts, and I grew up during the Big Dig. And when I say I grew up during the Big Dig, I mean it took 20 years, and I grew up during the Big Dig. And it's a really, it's an insane project. There's a there's actually an amazing podcast about it, which I would highly recommend to anybody. We

Albert Brier: 39:38

we recommend it regularly on the show.

Unknown: 39:40

Amazing, I love it. What projects are today, and what projects will be in the future? How they will be more integrated systems working together? The easy answer for me there is the small modular reactor program, particularly in the UK, because as you can guess, I am from there, and I spend a little bit of time working with. Rolls Royce on their SMR program about six or seven years ago. Seeing that shift happen, where the construction industry is being more technology focused and also really leveraging it and trying to become as efficient as possible, is really fascinating to me because I think if we have a better built world, right, everyone profits from that. I mean, right now it's the major bridges P3, so it's billion and a half dollar P3 initiative for PennDOT, replacing six interstate bridges, and so we've got four on I80, one on 78, and one on 81. Probably the Wesleyville site where Ontario Power Generation is potentially building up to 10,000 megawatts of nuclear power. It's it's been a while since anybody spend build a large scale nuclear facility. So that's exciting.

Nate Habermeyer, APR: 40:50

So what do you think? What was your first and or your big takeaway from that? Because it's super interesting.

Albert Brier: 40:57

I did think it was interesting that people mostly talked about their own projects.

Nate Habermeyer, APR: 41:00

That's right. It

Albert Brier: 41:02

you know it takes a particular kind of person to want to show up for a week long convention in for a lot of them another country, and those people are the most passionate, most dedicated folks in the industry in a lot of cases. So it makes sense that they would be passionate about their own projects. Maybe feel a little guilty, honestly, because there's like a 0% chance I would pick one of my own projects, like not not because I don't love my own projects. Of course, I do. They're all my special children, but there are so many other cool projects out there to talk about.

Nate Habermeyer, APR: 41:30

That's right.

Albert Brier: 41:30

And some people did right. Like there was there was one answer that really stuck out from the crowd.

Nate Habermeyer, APR: 41:34

It was interesting. Like it was really cool to hear the pride that people have in their work.

Albert Brier: 41:39

It really was, and

Nate Habermeyer, APR: 41:39

the passion, and that makes me realize there people really care about these huge endeavors that the general public is not necessarily very aware of, right? So that was pretty cool.

Albert Brier: 41:55

Yeah, especially since you would think that familiarity would breed contempt in this case, right? I I've been living and breathing this project for years. It's been a really difficult one. It has taken me away from my family. It has made my life more challenging. And you could be resentful of that, right?

Nate Habermeyer, APR: 42:10

That's right.

Albert Brier: 42:11

I've never felt that way about a project that I'm on, even when it has done all of those things. Like in the moment, sometimes I get really upset about it. But in retrospect, I always I learned something. I built a cool thing. Like you know. In some cases, I can go physically go to the thing that I built and look at it, be like, "Oh, that's cool. That's the thing that I did. Right?

Nate Habermeyer, APR: 42:27

Yeah.

Albert Brier: 42:27

That's a really special feeling that not everybody gets to feel. So it's nice to see that reflected in their answers. Like people really feel that pride. It was cool.

Nate Habermeyer, APR: 42:35

You were surrounded by a lot of experience. Yeah. And the people that you interviewed already, we've heard some really interesting hot takes on you know X Y and Z.

Albert Brier: 42:44

Yeah, well, I mean, you can really hear it from the the project experience that we just listened to. Like these people really know what they're talking about. So I asked them, what is that one big lesson that you learned early on in your career that you bring to all of your projects now? So let's let's listen to what folks had to say.

Unknown: 43:00

We have not been learning from our past mistakes. Our industry has been a laggard in adopting technology. There are instances where, even even in my own company and my own work with my colleagues, we will communicate a message one time and think that that message was received and understood. In reality, you need multiple times of repetition to have a chance of that being understood, all projects go over budget, and they all go over schedule. I want to listen. I really want to hear what's happening. I don't want to get right to the work. I don't want to like, hey, I've got the solution. Let's do it. I want to hear all the nuance because I think the nuance is where the story is. If you are bought into a schedule as an operational document, not a contractual document, it becomes something that your whole or organization operates around, and it's it's the key to delivering a project on time and on budget. First thing is, I would say, be on the job site if you're if you're a field person, be on the job site early. Building that trust with your staff is very important. You know, one lesson I learned early on in my career is that we're all going to be stretched, right? I think that applies to any project or any endeavor that we are asked to be involved in. But it's it's important not to allow ourselves or others in that stretching like a rubber band, right? If you think about it, not to let the rubber bands snap. You can have the most beautiful gold-plated project plan, but it ain't gonna roll that way. You've just got to be able to flex and change. I think knowing your team, see how to unlock their excitement and motivation. They are all not only working on a project to deliver, but also you need to know them, help them through their journey. I think basically everything comes down to people and communication. Listening to understand what the true challenge is, uncovering what those root causes are, what are symptoms, what's really the challenge that's happening, and. And how can the team work through it? As Ted Lasser says, be curious, not judgmental. And I found in every situation, every person I've met, every project I've worked on, and every country around the world I visited, over 30 now, if you're curious, you will learn something. Your goal should be making sure that the customer sees value, because the instant instant that the customer sees value, the renewal and the expansion is a byproduct. Don't be afraid to say that you screwed up. Right, the sooner you get it out there, and right, the sooner you try to fix it, the better off everybody is. Projects now is heavily dependent on the input from our the indigenous communities, but the but the problem is we're still learning about their culture, how to support their capacity, and how to work with them because a lot of them may not have a very you know for example for nuclear have a deep understanding how nuclear works. So basically, how to work together collaboratively.

Nate Habermeyer, APR: 46:00

Wow, pretty cool.

Albert Brier: 46:01

It was. It was. It was a lot of the same answer as the secret sauce. I found a lot of communication, interpersonal connection stuff. So that really was the theme of the discussion. What's your secret sauce? What's your lesson learned? Like, be a good communicator. It's pretty important, you know.

Nate Habermeyer, APR: 46:18

I did like. There's the bit about you know being on time, being on site early, being you know, oh yeah, man, that that one really spoke to me personally. That

Albert Brier: 46:27

that that one really did. I I I appreciated hearing that answer.

Nate Habermeyer, APR: 46:31

Yeah, that was really cool. So it sounds like the experience was just awesome, and I'm sad that I missed it. But I'm glad that you got some really cool interviews, which we will post in their entirety at a later point in time.

Albert Brier: 46:48

That's right

Nate Habermeyer, APR: 46:49

after this episode, so you can you know listen to everybody's the the full conversation with Albert on the floor. A lot of great interviews. Sounds like you had a lot of fun on the floor. I did, really did. Like a lot of really interesting perspectives and feedback from the people that you talked to. Anything stick out for you? Trends, like highlights that I think you know. Last, yeah, you know,

Albert Brier: 47:16

I think I think I know what you're trying to say. I think you're you're getting that. Like, what was what was my key takeaway from all that, right?

Unknown: 47:22

Yeah,

Albert Brier: 47:22

and it was really the the two things that we highlighted most frequently in the in the in the discussion just now. One, the importance of interpersonal skills and communication cannot be overstated. You can be as good as you want to be at the technical parts of your job and still be functionally useless because you can't communicate with other people. So this is a skill that you know I do not naturally possess a need to work on with people. So it really stuck out to me how many people say like, "Hey, this is this is my thing, right? This is what I do. It's like, well, that's aspirational for me. So it was nice to hear that because it it tells me I need I have a lot of a lot of work to do. But it's not just about me, right? It's about like how we train our junior people, right? That's one of my responsibilities here at Docanish. So we need to be training them not just to be excellent on their tools, but also excellent in a meeting room.

Nate Habermeyer, APR: 48:10

Yeah,

Albert Brier: 48:10

and a lot of that just comes from mentorship, right? Like letting them stick, you know, tag along with more experienced folks who go and deliver a presentation or have a conversation with a with a client or things like that. It's it's an important part of their growth and development. It shouldn't just be an also ran with the technical training. It needs to be part of the training. So that was one of my takeaways. And the other one was we really need to get off of our hands when it comes to implementing AI as an organization. We need to figure this out. Like we've got so many amazing tools out there doing so much awesome stuff that nobody knows about, right?

Nate Habermeyer, APR: 48:43

Yeah.

Albert Brier: 48:43

Why? You know, like we're we're carpenters in a room full of hammers, and we're still pounding nails in with our fists. Like, yeah,

Nate Habermeyer, APR: 48:52

I think there's a huge awareness, like education piece to to that. Just to start changing hearts and minds about these tools.

Albert Brier: 49:00

Yeah, and and also if we don't want like this is the other thing. If people feel like they're being pulled around by the nose ring about like AI and you will use it, like here's what it's going to do. Like you don't want it to be that way. Fine, tell people what you do want from it, right? Right. Like become a power user, get really familiar with all the stuff that it can do, and start being the person who makes those rules instead of somebody who just has to follow them. Like that's that's what I'm trying to do. Is like I want to generate enough. Like you know, we talked to a company called Construct Mind that builds schedules from scratch, and talking to those guys off mic about how much effort it took to to bring data in and like have it condition it and like how important the human in the loop part of it was, and designing that into the workflow and stuff like that. If you try to use a general purpose model for something like that, it's just not going to work, and that's why that's what I think is driving a lot of people when they feel that it's not working for them, or that it's it's useless, or can't do the things that people say it can do. It's because they're they're not aware of how. Much effort it takes to condition it to do things well

Nate Habermeyer, APR: 50:03

right

Albert Brier: 50:03

you know

Nate Habermeyer, APR: 50:04

right

Albert Brier: 50:04

so let's get off our hands and let's let's figure that stuff out

Nate Habermeyer, APR: 50:07

let's do it there you go people that's the that's the call to action well thank you Albert

Albert Brier: 50:12

yeah man thanks Nate this was a fun one

Nate Habermeyer, APR: 50:14

thanks everybody talk to you later

Albert Brier: 50:18

thanks for tuning in to the Risky Planner podcast, and thank you for your patience. It took a long time to turn this one around, but I hope you'll agree with me that it was really worth the wait. Particular thanks go out to the folks I interviewed at the AACE conference in June. Those folks are Giresh Batia from ConstructMind, Justin Jacobson from MBP, Matt Schonert from Hatch, Michael Lapage from Plan Academy, Philip Talbot from Nodes and Links, Prakesh Hundecker from Volkert, Barrett Richards from Turner and Townsend, Belinda Hussey from Transpower New Zealand, Diana Nada from Turner and Townsend, Holly Parkus from SMA Consulting, Rachel Fleming from MBP, Reese Phillips from Nplan, Rohit Sinda from Smart PM, Sarah Horsey from Bridging Pennsylvania Constructors and Tracy Leung from OPG. You guys made that experience really special for me. Thank you so much for your insights and for sharing your time. And I hope you enjoyed listening back to the interviews. The full interviews will come out as their own episode very very soon. Hopefully much sooner than last time. So look out for that. In the meantime, I hope you're enjoying our brand new theme music from Thompson Egbo Igbo. Head over to egbomusick.com. That's egbomusick.com if you'd like to hear more. While you're at it, you could leave us a review on your podcast platform of choice. Check us out on thereiskyplanner.com and share our stuff out to the folks you care about. I hope you found this one as special as I did, and I'll see you next time.